[
https://issues.apache.org/jira/browse/CASSANDRA-270?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jonathan Ellis updated CASSANDRA-270:
-------------------------------------
Attachment: 270-v2.txt
v2 updates xml to yaml Converter and makes deeper changes to BRAF: seeking back
to the current end-of-file also caused flush unnecessarily, since seek was in
charge of reloading the buffer for both reads and writes. moved that code into
reBuffer method, and made criteria different on writes (have to be writing at >
hi_) than reads (where we use >= hi as before).
Stu points out that this doesn't necessarily avoid seeks entirely -- it only
limits them to at most one per InMemoryCompactionLimit's worth, if a single row
crosses the buffer boundary. i think we can live with that.
> Reduce copies in data write path
> --------------------------------
>
> Key: CASSANDRA-270
> URL: https://issues.apache.org/jira/browse/CASSANDRA-270
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Todd Lipcon
> Fix For: 0.7
>
> Attachments: 270-v2.txt, 270.txt, patches.tar
>
>
> This is a series of patches against a very old version of Cassandra - they
> certainly won't apply, but Jonathan asked me to upload the patches here to do
> the ASF grant.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
